Adding Speckle-Tracking Echocardiography to Visual Assessment of Systolic Wall Motion Abnormalities Improves the Detection of Myocardial Infarction

Abstract

The aim of this study was to investigate whether speckle-tracking echocardiography (STE) improves the detection of myocardial infarction (MI) over visual assessment of systolic wall motion abnormalities (SWMAs) using delayed enhancement cardiac magnetic resonance imaging as a reference. Transthoracic echocardiography was performed in 95 patients with first ST segment elevation MI 110days (interquartile range, 97-171days) after MI and in 48 healthy control subjects. Two experienced observers independently assessed SWMAs. Separately, longitudinal peak negative, peak systolic, end-systolic, global strain, and strain rate were measured and averaged for the American Heart Association-recommended coronary artery perfusion territories. Receiver operating characteristic analysis was used to determine a single optimal cutoff value for each strain parameter. The diagnostic accuracy of an algorithm combining visual assessment and STE was evaluated. Median infarct size and transmurality were 15% (interquartile range, 7%-24%) and 64% (interquartile range, 46%-78%), respectively. Sensitivity, specificity, and accuracy of visual assessment to detect MI were 74% (95% CI, 63%-82%), 85% (95% CI, 72%-93%), and 78% (95% CI, 70%-84%), respectively. Among the strain parameters, SR had the highest diagnostic accuracy (area under the curve, 0.88; 95% CI, 0.83-0.94; cutoff value, -0.97sec-1). The combination with STE improved sensitivity compared with visual assessment alone (94%; 95% CI, 86%-97%; P<.001), minimally affecting specificity (79%; 95% CI, 65%-89%; P=.607). Overall accuracy improved to 89% (95% CI, 82%-93%; P=.011). Multivariate analysis accounting for age and sex demonstrated that SR was independently associated with MI (odds ratio, 2.0; 95% CI, 1.6-2.7). The sensitivity and diagnostic accuracy of visually detecting chronic MI by assessing SWMAs are moderate but substantially improve when adding STE.